United Kingdom Concludes Its Fifth Multi-Annual Spending Round

2007 Comprehensive Spending Review

Posted By Richard Hughes

In October of last year, the UK Treasury published the conclusions of its latest multi-annual expenditure planning exercise.  The 2007 Comprehensive Spending Review (CSR07) set out how the UK Government plans to spend its £600bn (US$1.2 trillion) annual budget over the years 2008-09, 2009-10 and 2010-11 and the key outcomes it is looking to buy with taxpayers’ money over the period. 

While the UK’s fiscal and budgetary architecture has been stable for over a decade and biennial Spending Reviews have become a routine feature of the British political calendar, there were are number of noteworthy developments and innovations in this fifth multi-year spending round.  In particular, CSR07:

• took place within a more challenging macro-fiscal context than previous spending rounds and was a test of the capacity of the UK’s Spending Review process to deliver an expenditure-based fiscal consolidation;
• saw a major expansion in the length, breadth and depth of budgetary certainty that the UK system provides to frontline managers; and
• included a major streamlining of the UK’s performance management regime for public services.

The UK’s Fiscal and Budgetary Framework

Before getting onto what was new this time around, a quick bit of background on the fiscal and budgetary framework put in place by the incoming Labour Government in 1997.

At the macro-level, tax and spending policy in the UK is anchored by two medium-term fiscal rules:

• a golden rule which requires the Government to keep the current budget in balance or surplus over the economic cycle; and
• a sustainable investment rule which requires the Government to keep public sector net debt below 40 percent of GDP.

To underpin the achievement of these medium-term objectives, strengthen the link between the new Government’s priorities and the allocation of public spending and facilitate longer-term planning on the part of departments (line ministries), the regime of annual “incremental” Public Expenditure Surveys which the UK had operated for three decades was replaced with a biennial cycle of “comprehensive” Spending Reviews undertaken every other year starting in the summer of 1998.

Spending Reviews are a detailed, bottom-up examination of each department’s budgetary requirements for the coming three-year period in light of existing spending pressures, opportunities for improving efficiency, and the costs of new policy proposals.  The review process is coordinated by the Treasury but led by departments themselves (though sometimes with input from independent experts) who submit detailed three-year resource requests to the Treasury three months prior to the conclusion of the exercise.

Spending Reviews are not, however, a purely bottom-up exercise.  The preparation and assessment of departments’ resource requests takes place within two global expenditure “envelopes” for total current and capital expenditure by the public sector.  These top-down constraints are fixed in the preceding Budget in March so as to ensure that the resulting spending plans are consistent with meeting the Government’s fiscal rules over the medium-term.  Within those global envelopes, total public expenditure is classified into two categories for planning purposes:

• Annually Managed Expenditure (AME) which is deemed too volatile, demand-led or sensitive to the economic cycle to plan on a multi-year basis.  This category includes debt interest, social security and a number of smaller items.  While the Spending Review includes a bottom-up forecast of the deferent components of AME, it does not fix any nominal limits for this category of expenditure and forecasts are updated in subsequent Budgets and Pre-Budget Reports; and
• Departmental Expenditure Limits (DEL) which covers the bulk of conventional public services such as health, education, transport, policing and defense for which the Government sets three-year nominal expenditure ceilings split into current and capital expenditure.  These ceilings impose a Treasury-enforced limit on each department’s parliamentary appropriations in those years and are generally only adjusted between Spending Reviews to reflect classification changes or unforeseen contingencies (for which the Government sets aside a small DEL Reserve).

A full forecast of Annually Managed Expenditure and 3-year Departmental Expenditure Limits for the 25 main UK Government departments are set out in the Spending Review White Paper which is presented to Parliament in the summer or early autumn of a Spending Review year.

These expenditure plans are accompanied by multi-year Public Service Agreements (PSAs), which articulate the key outcome-based performance objectives that the Government is committed to achieve with that expenditure over the three-year period.  Since coming to power in 1997, the Government has conducted five spending reviews in 1998, 2000, 2002, 2004 and 2007.

What was new in the 2007 Comprehensive Spending Review?

So, what was unique about this spending round by comparison with the four previous exercises?

A more challenging fiscal environment

The first new development was that the 2007 CSR represented the first real test of the capacity of the Spending Review process to plan and deliver a discretionary fiscal consolidation in the UK.  The previous four Spending Reviews took place within a context of an overall public expenditure envelope that was steadily growing as a share of the economy from 37% in 1999-00 to 42% by 2007-08.   

However, as both the UK’s fiscal rules began to bite, the UK needed to halve the real rate of growth in public spending from 4% p.a. over the last decade to 2% p.a. over the next three years – a ½% below than the trend rate of growth of the economy.  This is, of course, a modest fiscal consolidation by international (and indeed UK historical) standards.  However, it nonetheless posed a challenge for a Spending Review process which had been used to having a rapidly growing pie to divide between competing priorities.   

How did the Spending Review process measure up to the challenge?  There is a tendency for budget allocations to converge during a period of consolidation as Government’s favor sharing the burden more or less equally across line ministries above more strategic considerations.  While it was inevitable that the larger services such as health, education and local government would need to see some slowdown in the growth of their budgets in line with the total, it is notable that:

• Despite a halving in the real increase in overall resources available over the CSR07 period, some departments actually fared better in this spending round than in the last (notably those responsible for overseas aid, the intelligence agencies, defence and the environment) in recognition of their instrumental role in responding to the “global challenges” such as poverty, conflict and climate change that were an explicit theme of the 2007 CSR;
• Furthermore, the spread in real growth rates between these big winners and the biggest losers from the exercise actually increased (social security and tax administration and the Treasury itself) from 12 percentage points in the 2004 Spending Review to 16 percentage points this time around;
• Despite an even more dramatic slowdown in overall public sector net investment from the double digit real growth rates seen in past spending rounds to 4.5% p.a. over the CSR07 period, the Government managed to finance within this envelope two of the largest infrastructure projects in UK history.  Both in London, they are: (i) the five-year, £9.3bn (US$18.5bn) project for the construction of the park, venues and facilities for the 2012 Olympic Games in the city’s East End; and (ii) - the eight-year, £16bn (US$32bn) project for a new high-speed, cross-town rail service known as Crossrail which will link Heathrow Airport in the west to the growing commercial and residential centers in the Thames estuary and is due to begin service in 2017.
• Finally, in the face of competing demands for increasingly scarce resources, the Treasury managed to increase the size of its central contingency Reserve by £1bn to around 1% of total DEL by 2010-11 in recognition of the increased pressures on the Government’s expenditure plans in this more tightly-constrained environment.

So anecdotal evidence would seem to suggest that despite the slowdown in overall resources available, the UK’s comprehensive, bottom-up approach to multi-year expenditure planning managed to keep the focus on prioritization and prudence.

A major extension of multi-year certainty

A second noteworthy development in the 2007 CSR was a marked extension in the length, breadth and depth of certainty that the UK system now provides to managers about their future budgets. Taking each of these in turn:

• The length of budget certainty: The first four Spending Reviews between 1998 and 2004 were on a biennial cycle in which DELs were set for three-years but reviewed every two, so that the final year of the preceding exercise remained indicative and served as the first year of the next.  The 2007 CSR marked a break with this tradition by leaving the final year of the preceding Spending Review period (2007-08) fixed and setting new budgets for the subsequent three years (2008-09, 2009-10 and  2010-11) with no overlapping year between the two exercises.  As such CSR07 marked the move to a cycle of three-year fixed expenditure limits reviewed every three years.
• The breadth of budget certainty:  At the time of the first Comprehensive Spending Review in 1998, the split of total public expenditure between Annually Managed Expenditure and three-year Departmental Expenditure Limits was roughly 50/50.  Since then the Treasury has sought to progressively increase the proportion of expenditure included within DEL.  The 2007 CSR was preceded by a further such reclassification exercise which took the DEL/AME ratio up to 60/40 for the coming three years.
• The depth of budget certainty:  As Departments became more comfortable with operating within multi-annual budgets themselves, the Treasury has encouraged them to cascade that certainty to their delivery agencies.  This cascading process began in 2003 with the introduction of three-year budgets for each of the UK’s 300 local primary care heath trusts.   The 2007 CSR took this two steps further with the establishment of three-year budgets for every local authority and every school in the country.

Taken together, these three developments mean that from April of this year, about a third of public spending in the UK will be transmitted directly to frontline managers as guaranteed budgets for the next three years.

A further (r)evolution in the PSA framework

Finally, the 2007 Comprehensive Spending Review saw a major streamlining of the UK’s public service performance management regime in which 110 largely departmental-based Public Service Agreements were consolidated into 30 explicitly inter-departmental PSAs articulating the Government’s top priorities for the coming period such as:

• increasing adult literacy and numeracy;
• boosting the overall rate of adult employment
• halving child poverty; and
• raising the long-term supply of housing.

Each of the 30 new model PSAs was placed under the purview of an interdepartmental Cabinet committee and underpinned by a published delivery agreement setting out the role of each contributing department in delivering the agreed outcomes for the next three year period. 

This slimming down of the number of objectives set by the center was coupled with a clearing away of the undergrowth of over 1200 performance indicators that had grown-up at the local level.  From 2008, local authorities in England will only be required to report against 198 performance indicators of which a maximum of 52 can be set by central government departments.  The remainder will be determined at the local level through Local Area Agreements drawn up between elected councils and local partners.

Has the UK Spending Review model stood the test of time?

A decade on from the first Comprehensive Spending Review, the 2007 CSR demonstrated how the UK’s “top-down meets bottom-up” approach to multi-year expenditure planning has continued to develop and adapt to the changing demands being placed upon it by macroeconomic and political factors.  However, fixing public spending plans is the easy part - like the theory part of the driving test.  The real test of the resilience of the Spending Review process is whether the UK can stay within the yellow lines during the practical part of the exam between now and March of 2011…

Comments

[Anand P. Gupta]

I have read with a great interest the material on UK's comprehensive public spending review. It does not provide any information on the effectiveness of public spending in delivering the intended outcomes. Is public spending more effective today than it was, say, five years ago? If so, how did this happen? I will greatly appreciate detailed responses to these questions.

[Richard Hughes]

A straightforward question, though not an easy one to answer in a straightforward way - as it begs the question “more effective in doing what?”

If you mean “Is the Government now spending more on the services people care about and less on the services people don’t care about?”, well it’s difficult to find a public service that nobody cares about. But to govern is to choose, and the present Government was first elected in 1997 on a manifesto that promised to:

- boost the proportion of public spending that went on “priority public services” of education, health and transport; and

- reduce the proportion that went on “the costs of economic failure” such as debt interest and unemployment benefits.

These have been explicit objectives of successive Spending Reviews since the first in 1998. Judged against this standard, most would accept that the Government has achieved what it set out to achieve and that its medium-term oriented fiscal and budgetary reforms (backed up by specific policy reforms like the New Deal welfare-to-work program) were instrumental to that success. The proportion of total UK public expenditure going to the Government’s stated priorities of health, education and transport has risen from 28% in 1997 to 35% today while the proportion going on debt interest and unemployment has fallen from around 10% in 1997 to 6% today.

Another way of asking your question is, “Has this additional expenditure improved outcomes in the real world?” Here again, you can only judge Governments by what they set out to achieve at the beginning. Alongside promises to increase resources going to health, education and transport, the Government also committed itself in 1997 to a series of outcome-based performance objectives in each of these areas such as:

- reducing mortality rates from cancer and heart disease;

- increasing childhood literacy and numeracy rates; and

- improving the reliability and punctuality of rail services.

These objectives were codified in the Public Service Agreements or PSAs I referred to in my original posting. Notwithstanding the measurement challenges involved, again few observers would disagree that standards have improved in these areas:

- mortality rates have fallen by 14% for cancer and 31% for heart disease since 1997;

- literacy at age 11 has risen from 65% in 1997 to 80% today and numeracy at age 11 has risen from 59% in 1997 to 77% today; and

- rail reliability has increased from a low of 75% in 2001 to 85% today.

And most within the public sector would accept that PSAs played a major role in ensuring a sustained focus on delivering sustained improvements in these objectives (though there continues to be a debate about whether these were and are the right objectives to have).

A final (and much more tricky) formulation of your question is “Given the rate of increase in expenditure within each service area, is this rate of improvement in outcomes better than one might have expected?” or (more crudely) “Is the UK taxpayer now getting more punch for his or her pound?” Here we run up against the limits of what the available data can tell us (and into the centre of an ongoing debate about public services in the UK). Because most of the indicators of public service outcomes that exist in the UK were created for the purposes of monitoring progress against PSA targets, it is difficult to find a consistent time series that stretches back before 1998-99. And because the relationship between expenditure and outcomes is at best a “grey box” mediated by a range of environmental factors, it is difficult to specify a detailed production function for something like the rate of crime and then isolate the impact of expenditure on policing or other services.

That said, what one tends to observe in the relationship between expenditure and outputs in the UK over the ten years for which we have data is:

- a lag of a year or so as systems are reformed, new staff are hired and people get used to having more money to spend;

- a rapid improvement in outcomes for a number of years as that additional activity has an impact;

- a slowdown in the rate of improvement as easier cases are dealt with, tougher ones remain and initial energy and focus dissipates;

- a plateau or shallow upward trajectory at the new, higher level of performance.

All of which suggests that public services are not immune to the principle of diminishing marginal returns. And this brings me back to the point I made in my original posting about a major test of any system of expenditure allocation being its capacity to respond to information about falling returns in one area by focusing additional expenditure on new priorities. Here again, I think the UK’s “bottom-up meets top-down” approach does a reasonable job – as evinced by the reprioritization of expenditure toward historically less-favored departments in the most recent Spending Review.

[Anand P. Gupta]

Many thanks for your prompt response. Yes, there can be distortions in how a government allocates its resources – it may spend on the wrong goods or the wrong people. But the issue that I raised relates to the effectiveness of public spending in delivering the intended outcomes. The issue, to use your formulation, is: “Is the UK taxpayer now getting more punch for his or her pound?” I understand the conceptual and measurement issues that this poses. You refer to the difficulty of specifying a detailed production function for something like the rate of crime and then isolating the impact of expenditure on policing or other services. But this precisely is the point. What you say about the crime rate also applies to, say, infant mortality. How has UK dealt with such situations? How do you define accountability in such situations?